The p53 tumor suppressor is a DNA damage/stress response protein that functions as a transcription factor to regulate a large number of genes that prevent proliferation of damaged cells via initiation of cell cycle arrest and senescent programs or via apoptosis and other mechanisms of cell death which are potent tumor suppressive mechanisms. Disruption of the pathway occurs most often through mutation or deletion of the p53 gene itself, but elevated levels of two important p53 inhibitors, MDM2 and MDM4, also contribute to tumor development in specific cancers. MDM2 encodes an E3 ubiquitin ligase that targets p53 for degradation and both MDM2 and MDM4 inhibit p53 activity by masking its transcriptional activation domain. In mice, overexpression of Mdm2 or Mdm4 is sufficient to drive tumorigenesis, while deletion of either Mdm2 or Mdm4 results in p53-dependent lethal phenotypes. These and other studies provide compelling data for the relationship between Mdm proteins and p53 and have led to the development of MDM2 inhibitors (MDM2i) for the treatment of human cancers. Recently published Phase I clinical trials using one such MDM2i show promise as several patients achieved complete or partial responses or had stable disease following treatment and tumors exhibited activation of downstream p53 target genes. Unfortunately, these studies also highlighted serious adverse effects including hematological, gastrointestinal and kidney toxicities, phenotypes that were predicted from studies in mouse models. In addition, activation of known p53 targets did not always correlate with p53-mediated cellular events and implies tissue specificity of response. We have developed in vivo mouse models that allow us to probe the specificity of the p53 response at the molecular and organismal levels. Global loss of Mdm2 results in tissue-specific activation of p53 targets and apoptosis/senescence phenotypes which also vary by tissue type. In another model, deletion of the p53 target Puma but not p21 rescued some of the defects due to reduced MDM2 levels. These tissue specific differences need to be better defined to understand the toxicities of Mdm2i in humans and to develop potential biomarkers for toxicity. Thus the first aim will be to determine and functionally examine the p53 transcriptional program and the downstream pathways (senescence/apoptosis) that are activated in vivo upon depletion of Mdm2 in bone marrow, intestine and kidney. High MDM2 levels as occur in some human cancers are not tolerated by normal cells. We have completed a CRISPR/Cas9 screen to identify factors that allow normal cells to survive despite elevated levels of MDM2 to identify and characterize synthetic lethal relationships with high MDM2 in tumors (aim 2). Lastly, we plan to determine the mechanisms and factors that cause increased MDM2 levels in tumors that lack amplification of the Mdm2 locus since small differences in MDM2 levels (and therefore p53) impact tumorigenesis (aim 3). Our studies will deepen our understanding of the p53-Mdm2 pathway, yield possible markers for toxicities to MDM2i, and uncover the molecular mechanisms that induce and support MDM2 overexpression in cancers.